These studies have two overall goals: First, to further our understanding of protein-DNA interactions, particularly the role of DNA structure in site recognition, and second, to better understand the biological roles of the highly conserved HU/IHF family of proteins that have been selected for detailed study. Recognition of particular DNA sites by proteins is the sine qua non for the regulation of gene expression and many other biological processes. In some cases, such recognition is a straightforward consequence of hydrogen bonds between properly arrayed protein side chains and the edges of bases in the major groove of B-form DNA. In many instances, however, the situation is much more complicated, and recognition depends on sequence - dependent variations in the structure and distortability of the DNA ("indirect readout"). This form of recognition is not as well understood as the former. The IHF/HU family proteins are small, closely related prokaryotic DNA bending proteins that function as architectural factors in a variety of processes (e.g. transcription and recombination) that require multicomponent protein-DNA complexes. Two of the proteins to be studied, IHF and Hbb, recognize specific (yet different) sequences in the DNA, but do so almost entirely through indirect readout. The third, HU, binds nearly independently of DNA sequence but recognizes specific structural distortions in DNA, and may play a role in DNA repair. It is well established that IHF binding introduces a nearly 180 degree bend in the DNA, but estimates of the bend introduced by HU vary widely. This uncertainty hampers our understanding of how HU functions in conjunction with other proteins in the cell. This project will study the binding properties of these proteins both in solution and by x-ray crystallography. Hypotheses based on comparing and contrasting such data for the 3 different proteins to be studied will be tested by site-directed mutagenesis and domain-swap experiments.